|
Post by moneyball on Jan 12, 2015 20:38:57 GMT
Earlier today I noticed that there was a sizable (at least £160,000) order to borrow at 4.5% in the 3Yr Income market. What was confusing, was that there also appeared to be lenders offers (both for smaller amounts) outstanding at 4.3% and 4.4%.
At the time of writing these have disappeared (presumably matched with this large order or otherwise) but theres still £110k to be matched for the borrower and £161k to be matched for lenders - both at 4.5%.
Any ideas on why this is the case? As it would logically appear as roughly £51k outstanding for lenders and zero for the borrower.
For clarity, I have no funds awaiting to be matched in this market at present, Im asking through pure curiosity!
|
|
sl75
Posts: 2,092
Likes: 1,245
|
Post by sl75 on Jan 12, 2015 21:26:17 GMT
It certainly used to be the case that the "behind-the-scenes" matching process operated to 2 decimals of %, whereas the rates on the marketplace were only shown to 1 decimal of %.
As I understand it, borrowers and lenders would each see the rate that applies to them - so for a lender it's the return after all fees, and for the borrower it's the APR inclusive of all fees... so 4.5% to a lender might translate to (say) 10.32% for a borrower. If the borrower asks for 10.3% instead, the system would effectively translate that to a request at 4.48%, which also displays as 4.5%.
|
|
pikestaff
Member of DD Central
Posts: 2,136
Likes: 1,484
|
Post by pikestaff on Jan 12, 2015 23:19:40 GMT
It certainly used to be the case that the "behind-the-scenes" matching process operated to 2 decimals of %, whereas the rates on the marketplace were only shown to 1 decimal of %. As I understand it, borrowers and lenders would each see the rate that applies to them - so for a lender it's the return after all fees, and for the borrower it's the APR inclusive of all fees... so 4.5% to a lender might translate to (say) 10.32% for a borrower. If the borrower asks for 10.3% instead, the system would effectively translate that to a request at 4.48%, which also displays as 4.5%. This was given as the explanation for a previous anomaly but I do not believe it to be correct p2pindependentforum.com/post/26349/threadp2pindependentforum.com/post/26551/thread
|
|
|
Post by moneyball on Jan 13, 2015 18:29:58 GMT
It certainly used to be the case that the "behind-the-scenes" matching process operated to 2 decimals of %, whereas the rates on the marketplace were only shown to 1 decimal of %. As I understand it, borrowers and lenders would each see the rate that applies to them - so for a lender it's the return after all fees, and for the borrower it's the APR inclusive of all fees... so 4.5% to a lender might translate to (say) 10.32% for a borrower. If the borrower asks for 10.3% instead, the system would effectively translate that to a request at 4.48%, which also displays as 4.5%. Thanks for the response. That scenario did cross my mind too, however, if true, it wouldnt explain the earlier situation of outstanding lender requests at 4.3% and 4.4%. Unless theres some further drilling down into the decimal places aspect that is required? My first thought was that I must be looking at and "old" version of the page, ie, a browser problem but that appears to be fine everywhere else and I tried it on two devices - the same info appeared on both. Could the fact that it was a sizable (especially for the 3Yr market) order have some weird consequence too?
|
|
sl75
Posts: 2,092
Likes: 1,245
|
Post by sl75 on Jan 13, 2015 18:53:06 GMT
Regarding the point you make in the second linked post, it would be expected that you would only be able to witness a different rate (to 2 d.p.) if you had "crossed" a borrower request rather than leaving funds on offer for a borrower to match later. I'd hypothesised in my subsequent post that a "crossing" offer for an amount larger than the borrower request could potentially leave a residual amount on offer at the borrower's requested rate (rather than at the normal rate for that 1 d.p. advertised rate), but as I recall, a different reason was found for that particular anomoly. The issue with unfulfilled buy/sell orders that *appear* to be at the same rate to 1 d.p., however, was sufficiently well-known in the past that RS had a banner that automatically appeared in order to explain it. That banner seems no longer to be present (or no longer triggered to appear) which either indicates that it got removed (perhaps inadvertently) in the cosmetic redesign of the site, or that the system behaviour has subsequently changed to avoid the "rates that seem the same but aren't" problem. If the latter case, then clearly the former explanation would no longer apply. moneyball, at the time I looked soon after your original post, there was unfulfilled borrower demand at 4.3% and 4.4%, not outstanding lender requests. It was several minutes later though, so what I saw may not have been exactly what you saw... I agree it would be very strange indeed for there to be lender requests that are actually lower rate than the unfulfilled borrower demand, and that the explanation of what I had seen would not explain that.
|
|
pikestaff
Member of DD Central
Posts: 2,136
Likes: 1,484
|
Post by pikestaff on Jan 13, 2015 19:43:47 GMT
...Regarding the point you make in the second linked post, it would be expected that you would only be able to witness a different rate (to 2 d.p.) if you had "crossed" a borrower request rather than leaving funds on offer for a borrower to match later... The point I was trying to make, but perhaps did not make clearly enough, is that although borrower rates are expressed to 2dp the set of rates available to borrowers is not all possible 2dp rates. There is a SINGLE borrower rate that corresponds to each 1 dp offered rate, and these are the only borrower rates that exist. Or so I have observed based on the entire population of my matched offers. An offer of (say) 5.8% corresponds to EXACTLY the same borrower rate on every day that I have offered at that rate, and this 1:1 relationship is true for EVERY rate at which I have bid. It's also true regardless of how long my money was on offer (anything from minutes to several days). This statement is not based on a 100% sample (that would be too boring) but it is based on a fairly random sample of over 50 matched bids, checking back to the loan documentation in each case.
|
|
sl75
Posts: 2,092
Likes: 1,245
|
Post by sl75 on Jan 13, 2015 23:37:27 GMT
An offer of (say) 5.8% corresponds to EXACTLY the same borrower rate on every day that I have offered at that rate, and this 1:1 relationship is true for EVERY rate at which I have bid. It's also true regardless of how long my money was on offer (anything from minutes to several days). I think we're talking cross purposes - my understanding (having not seen that part of the borrower experience) is that borrowers see the market as their APR. This includes the effect of that specific borrower's personal "credit rate" payment and the RS fee, and would be somewhat higher than the contract interest rate. I'm unclear on what rate you are referring to as the "borrower rate", as the one I'm referring to would not be directly visible to lenders anywhere I am aware of. This would usually be higher than the contract rate, or than the rate the lender receives. I assume the borrowers access the market to 1 d.p. too, but an exact rate like 5.80000% would NOT in general precisely correspond with a borrower APR of X.X0000%, and as the effect of the various fees will not be identical from one borrower to another, I would not expect a rate that a borrower directly enters to precisely correspond with the rate a different borrower enters, or the rate that a lender enters. Where a borrower matches a lender's existing offer already on the marketplace, it is the lender who sets the rate. However, it is also possible at RS for a lender to match a borrower's existing request already on the marketplace. If the market operates broadly as I'd previously understood it (which I'm not guaranteeing!), the rates that are set by the borrowers would not in general be aligned with the rates set by the lenders.
|
|
pikestaff
Member of DD Central
Posts: 2,136
Likes: 1,484
|
Post by pikestaff on Jan 14, 2015 10:07:59 GMT
sl75 The interest rate that I am referring to as the borrower rate is the actual interest rate to 2dp as shown in the contract. If you go to your money on loan you will see a list of contracts. Click on "view" for any one of those contracts and it will take you to a "loan contract details" page. Click on "view" against "contract document" and this will enable you to open or save a pdf of the contract. The contract has a 2dp interest rate at the top. Note that this is separate from the credit fee agreed between RS and the borrower. As i said in one of my earlier linked posts, for contracts entered into since RS stopped charging lender fees, my experience based on a sample of my loan book, originated on a range of different dates is that: all of the 5.9% loans have a contract rate of 5.75%, which is the minimum rate needed to round down, after compounding, to 5.9%. all of the 6.0% loans have a contract rate of 5.85%, which is the minimum rate needed to round down, after compounding, to 6.0%. The same is true for different rates. All of the loans that I have looked at have the minimum rate needed to round down, after compounding, to a specified single dp rate. This is what I would have expected. There is no good reason for the loans ever to have an intermediate rate. This relationship also holds true for older loans where there is a 10% lender fee, if it is deducted before compounding. All that said, I take your point that the 1:1 relationship might not apply where lenders match an existing borrower bid. But any departure from the 1:1 relationship would introduce inefficiency in the market because some borrowers would be paying more than is required to give lenders their stated 1dp rate. Either that or some lenders would be receiving less than their stated rate, which I hope very much is not the case! In the interest of science I have just matched a small amount against an existing borrower bid of 5.6% in the 5 year market. It was snapped up immediately. When the loan has been formed I will look at the contract and report back.
|
|
sl75
Posts: 2,092
Likes: 1,245
|
Post by sl75 on Jan 14, 2015 10:23:21 GMT
All that said, I take your point that the 1:1 relationship might not apply where lenders match an existing borrower bid. But any departure from the 1:1 relationship would introduce inefficiency in the market because some borrowers would be paying more than is necessary to pay in order to give lenders a given 1dp rate. When you switch the market the other way, the same inefficiency appears with lender-set rates from a borrower's perspective - a borrower satisfied with a given 1dp APR would be paying less than they are prepared to if they match an existing lender offer (as in general the rates set by lenders would not be nicely aligned with a round value as a borrower's APR). If that understanding of the marketplace is in fact correct, RS could make it more transparent (but also more confusing) by displaying 2 d.p. rates, but still only letting you specify 1 d.p. (so lenders still need to undercut other lenders' offers by at least 10 basis points, but could choose to match a crossing borrower request that is not as much as 10 basis points below the other lenders' offers).
|
|
pikestaff
Member of DD Central
Posts: 2,136
Likes: 1,484
|
Post by pikestaff on Jan 14, 2015 10:38:32 GMT
sl75 I'm not a borrower so I'm not sure if their site works exactly the same way. However, if they are specifying annualised rates to 1 dp I think they should in principle correspond to the same limited set of 2 dp rates that I have observed in my contracts and thus to the same set of annualised 1 dp rates that we bid as lenders, with no rounding differences whatsoever. A difference would, IMO, be indicative of a programming error, which is what I think happened the last time there was an anomaly. Probably someody who did not do the original coding rounding something to the wrong number of decimal places. It's just occured to me that if that is the problem my "scientific" test is unlikely to show it up. I may have cost myself 10p (if the contract runs to term) in vain! Let's wait and see what the contract says anyway...
|
|
leonp
New Member
Posts: 2
|
Post by leonp on Jan 14, 2015 11:06:53 GMT
Hi all,
I'm new here - reading for a while and think this is my first post!
I too noticed this and thought it odd. There was lender money available at 4.3, 4.4 and 4.5%, the total of which exceeded the the amount a borrower had requested at 4.5%.
It occured to me at the time that it was odd that the money hadn't been matched.
All of that said, 10 minutes earlier I received an email informing me of a successful referral and that my £25 had been placed in the monthly market. I managed to claw it back from the monthly market and added an offer at 4.5% in the 3YR market.
The next page I saw, informed me that my money had already been matched (rather than telling me my money was now offered).
It took a while after I had offered that money (and been told it had been matched) before it appeared to sort itself out in the market view.
Leon
|
|
sl75
Posts: 2,092
Likes: 1,245
|
Post by sl75 on Jan 14, 2015 12:02:52 GMT
sl75 I'm not a borrower so I'm not sure if their site works exactly the same way. However, if they are specifying annualised rates to 1 dp I think they should in principle correspond to the same limited set of 2 dp rates that I have observed in my contracts and thus to the same set of annualised 1 dp rates that we bid as lenders, with no rounding differences whatsoever. That won't usually be the case. A lender's annualised rate is net of all fees, and a borrower's annualised rate includes all fees (one of which is the borrower's personalised "credit rate" payment, so different borrowers should see the same market as though it has very different numbers - different both to what lenders see and to what each other see). It would be unlikely for both to be "nice round numbers" to 1 d.p. at the same time. I did try the first step of an application recently. I was quoted an APR of 11.1% from the initial check, which wasn't worth it for my purposes at the moment (I'll just draw down from P2P accounts, and if necessary use Zopa's Rapid Return), so didn't bother proceeding further (thereby wasting RateSetter staff time for any further checks they'd do in order to approve me to borrow at a rate I wasn't interested in borrowing at anyway). If my requirement had been more urgent, or I'd otherwise decided to take it further, I might have been given the opportunity to attempt to interact with the marketplace (I didn't, so still have no idea exactly what user interface or interactions they have for the borrower-side of the marketplace) - I might have tried for an APR of 9.9% (to get below the psychological 10% barrier), in which case lenders would have seen a demand at a rate something in the region of 1% lower than the prevailing market rate... but being calculated from MY request for 9.9% APR, it would be unlikely to be exactly correspond with the relevant round number in terms of lender return.
|
|
sl75
Posts: 2,092
Likes: 1,245
|
Post by sl75 on Jan 14, 2015 14:16:36 GMT
It's just occured to me that if that is the problem my "scientific" test is unlikely to show it up. I may have cost myself 10p (if the contract runs to term) in vain! Let's wait and see what the contract says anyway... It'll be interesting to see the outcome of your test, however, in the interest of scientific enquiry I also tried my own. In order to minimise the overall cost of this test, I did it in the "monthly" market with £10 stakes. First, I matched an otherwise unmatched borrower request which displayed as 2.6% in my view of the market as a lender. The resulting contract showed an interest rate of 2.65%. Second, I waited until there were no unmatched borrower requests, and placed £10 at a lender-requested rate of 2.6%. This was later matched by a borrower, and the resulting contract shows an interest rate of 2.57%. In practice the difference in rates is irrelevant for a £10 stake over that period, as both contracts will be accruing exactly £0.02 of interest during their 1 month lifetime (which looks to me like an annual return of only 2.4%!) Indeed, it may be possible to carefully calculate an offer size that, based on RS's rounding methods, would maximise the return - e.g. the minimum offer that would generate £0.03 of interest... but further research would be required, perhaps sponsored by someone interested in that result.
|
|
pikestaff
Member of DD Central
Posts: 2,136
Likes: 1,484
|
Post by pikestaff on Jan 14, 2015 15:20:55 GMT
sl75 I now have the result of my test too and you are right. I matched a borrower request that was displaying at 5.6% in the 5 year market and it has come back with a contract rate of 5.54%, which corresponds to an annualised rate of (1+5.54%/12)^12-1 = 5.683% whereas if the borrower had matched a lender request at 5.6% the contract rate would have been 5.47% (being the smallest 2dp rate that compounds to over 5.6%). It's disappointing to be proved wrong but the good news is the test should end up costing me about 3p less than I expected!
|
|
gt94sss2
Member of DD Central
Posts: 281
Likes: 137
|
Post by gt94sss2 on Jan 14, 2015 16:18:40 GMT
I'm new here - reading for a while and think this is my first post! I too noticed this and thought it odd. There was lender money available at 4.3, 4.4 and 4.5%, the total of which exceeded the the amount a borrower had requested at 4.5%. It occured to me at the time that it was odd that the money hadn't been matched. I'm new here and to RS as well - and glad someone asked this as well. I had the same experience as you in matching that large order. I still don't really understand why the funds already on the market weren't matched - mist reread this thread! Would also be interested in knowing more about the comments I have seen about ''My rate' not working as one would expect'
|
|